Brain dynamic low-frequency (<1Hz) fluctuation (LFF) signal has been observed using different brain imaging modalities cross multiple scales. For example, the large-scale whole brain resting-state (rs) fMRI signal fluctua- tion (<0.1Hz) has been linked to specific brain states, e.g. sleep, arousal switches, and anesthetic states. Animal fMRI has played a critical role in mapping brain function with multi-modality techniques to link the LFF temporal features across different spatial scales. The LFF correlation feature has been reported with concurrent fMRI and electrophysiology or calcium recordings at the single cortical site or through wide-field optical imaging methods at the cortical surface. However, the large-scale functional connectivity has not been well interpreted at the cellular level with laminar specificity or through subcortical/cortico-cortical projections. This gap of knowledge is primarily due to the lack of technologies to detect the multi-site laminar-specific fMRI LFF with concurrent cell- specific neurovascular signaling events at different cortical layers in animals. The goal of this proposal is to build an advanced multi-modal fMRI platform to study the laminar-specific LFF in awake animal models. We will merge the advanced fMRI methods, e.g. the line-scanning fMRI, based on a novel design of Switchable, Wireless, Implantable RF coil Array (SWIRFA) with a photonic crystal fiber (PCF)-based multi-channel MR-compatible laminar-specific imaging device for intracellular calcium (Ca2+ sensor: GCaMP6f) and extracellular Glutamate (Glu sensor: GluSnRf) recordings in awake rodents. The SWIRFA will be mounted with the head-post above the skull to reduce the distance between the RF coil and the cortex so as to boost the SNR for the line-scanning based laminar rs-fMRI with a >10 Hz fast sampling rate. Using the PCF array with beveled tips, the layer-specific Ca2+/Glu signals of individual cells can be recorded from multi-channels by parallel beam projection to a fast Silicon Photomultiplier (SiPM) sensor array. Three aims will be addressed: 1). Develop the Multi-site line scan- ning fMRI: Implement the SWIRFA for multi-slice line-scanning fMRI. We will apply the Wireless Amplified Nu- clear MR Detector (WAND) scheme to build a coil-array for switchable and wireless parallel RF signal transmis- sion and develop the multi-site line scanning methods for laminar-specific rs-fMRI. 2) Establish the Laminar- specific Ca2+/Glu recording: Develop a PCF-based multi-channel single-cell recording device across different cortical layers with the ultrafast sampling rate. 3). Validate the Multi-modal fMRI platform in awake mice: Combine the SWIRFA-based line-scanning fMRI with PCF-based cellular recordings for laminar-specific multi- modal fMRI studies. Eventually, the integrated multi-modal fMRI approach provides key strategies to not only improve the mechanistic understanding of resting-state networks (RSNs) at the cellular and circuit levels in ani- mal models but also elucidate the LFF-specific neural correlates of fMRI in the human brain at varied states and validate the clinical indications of altered RSNs in the diseased brain.